Colorectal cancer (CRC) is the second leading cause of cancer death in the United States. Early detection of CRC through routine colonoscopy is critical to a favorable outcome, however there is growing evidence that tumor morphology can impact frequency of detection. Classical polypoid lesions are detected and removed during colonoscopy at a much higher rate than flat lesions, and there is growing evidence that once detected, flat lesions have a higher propensity to be invasive. Little is known about how or why different CRC morphologies form or whether it is due to specific genetic or environmental factors. Recently, novel mouse models of CRC that consistently and almost exclusively produce flat colorectal tumors have been developed. This suggests that there are specific genetic host factors that determine CRC morphology. Using these mouse models, we propose to identify the genetic factors responsible for CRC morphology by 1) determining the number and genetic locations of flat CRC modifiers and 2) fine mapping and identifying candidate genes for modifiers of flat CRCs. A genetic cross has been established between two mouse strains that develop contrasting CRC phenotypes. Genotypic and phenotypic data will be collected and a statistical genetic analysis of F2 and backcross mice will be performed to elucidate the genetic intervals containing non-polypoid associated modifiers. Concurrently, microarrays will be performed on colon tissue from the crosses and an eQTL study performed to identify modifier candidate genes.